We need more inflation because we need more AD
Mark Thoma has a new post over at CBSMoneywatch.com where he asks why so many economists are now criticizing Bernanke for failing to raise the Fed’s inflation target. At one point he challenges economists who do favor higher inflation to come up with a model that would justify this policy. Thoma doesn’t think that new Keynesian models featuring price stickiness are applicable to the current crisis, which he says is caused by banking problems, not price stickiness.
I think Thoma’s views are probably pretty widespread, but I also think his article perfectly illustrates how the subject of inflation targeting can lead economists astray. The answer to the question he raises is actually very simple; we need more inflation because we need more AD. Ah, you might be thinking, but how do you know we need more AD? My response is that Thoma seems to think we do, at least that’s the implication of statements like this:
I have emphasized a portfolio approach involving both monetary and fiscal policy in the hopes that one or the other will get the job done.
Indeed, unless you are a professor at a major salt water economics department, I would think that it is pretty obvious that the economy could use a bit more aggregate demand. [edit: I meant freshwater university.] But what does that have to do with inflation? Inflation is not a good thing in and of itself; rather it is an inevitable side effect of more AD. (OK, if the SRAS were completely flat you might not get any additional inflation. But that would be a situation where more monetary ease would be especially useful, so it’s hardly an argument against a more expansionary Fed.)
I think a lot of people tend to over-think the problem. They look for all sorts of transmission mechanisms. More inflation might lower real wage rates, or it might lower real interest rates, or it might boost velocity. But it is really much simpler than that. We need more inflation because we need more AD. In that case you might ask; “why target inflation then; why not target AD directly, if that is what you are trying to increase?”
That would be a great idea! Does anyone know of a good proxy for total spending on all goods and services in the economy? If we could find such a proxy, we could simply target the growth rate of that variable, and then forget about inflation. After all, it is spending we are trying to boost, not inflation. Any suggestions?
PS. The government just downgraded NGDP growth in the 3rd quarter from the original estimate of 4.4% to 2.6%. You may recall all the fiscal stimulus fans crowing about how the 4.4% figure showed stimulus was finally working, even though it was below trend. Now it seems like fiscal stimulus has failed, so obviously we need even more fiscal stimulus. I used to argue that every American would need to have their wage cut 8% relative to a trend line from mid-2008 to restore full employment at current levels of NGDP. I’ve never met anyone who has had that big a wage cut. With the new NGDP figures even an 8% cut relative to their normal increase wouldn’t get the job done. One of three things must happen:
1. The Fed needs to produce much higher NGDP.
2. Americans need to accept far deeper cuts in their wages. And not just factory workers, not just unemployment workers, everyone needs a deep pay cut.
3. We can endure years more of high unemployment, with many lives being ruined.
Something for the Fed to think about.
HT. Marcus
Tags:
22. December 2009 at 15:34
Do you mean “sweet water” there? — Around the lakes, Minn, Chicago, etc.
22. December 2009 at 16:29
Scott: “That would be a great idea! Does anyone know of a good proxy for total spending on all goods and services in the economy?”
Just to be clear, you do mean *nominal* spending, don’t you? (Because otherwise people might misunderstand you, and think you wanted to use monetary policy to target some real variable.)
Otherwise, a lovely clear post. It’s not that we want inflation, even as a means to an end. It’s just that we want something that would happen to cause inflation as a side-effect.
22. December 2009 at 17:12
Thanks Agent, I made the change.
Nick, Yes, I am always talking about NGDP targets so I figured that people would assume that was what I meant. I agree that targeting RGDP would be a mistake.
22. December 2009 at 17:12
I don’t think it would work, because of who gets the money first.
Growth in AS causes inflation, yes, but so does a drop in AS.
Explanation:
Because of certain structural issues, I believe that inflationary monetary policy would increase the price of inputs before it would increase the price of outputs. The higher input price would cut marginal profits which would then result in more reduced production. I believe at this point that expansionary monetary policy would lead to inflation and less positive RGDP growth, at this stage.
This would not be the case if instead of money going to banks it went directly to consumers, as with a tax cut, or if the money went to cutting inputs cost by cutting the employer portion of payroll taxes.
22. December 2009 at 17:23
Let’s see
Nominal expenditure?
I don’t think Toma is “overthinking” it. A higher target for inflation is a way to reduce the real federal funds rate, making it sufficiently negative so that real expenditures will rise.
Monetary policy can’t be anything other than an effort to manipulate the federal funds rate.
Our story is something like the Fed targets nominal expenditure to return to target. (Maybe 11% or whatever the revised figure will be is too much for one year, but something.)
Let’s stick with 11%. If real output doesn’t change at all, the inflation rate will be about 11%. If the Fed funds rate were .2%, that would be a negative fed funds rate rate of 10.8%. This would be sufficient to motivate large increases in nominal expenditures and an increase in the price level at least 11%.
On the other hand, suppose inflation continues at 1% and real output rises 10%. With strong expected real growth, thre is motivate to raise nominal expenditure–investment and consumption. The natural interest rate rises. Credit demand rises. The .2% federal funds rate is more that consistent with nominal expenditure rising enough for recovery.
Whatever the mix of price increases or real output increases, as long as nominal expenditures are expected to rise, there is motivation indviduals to spend more.
I thought it was odd that Toma identifies quantitative easing with creating expected inflation and so lowering the real fed funds rate (or some real interest rate.)
I think quantitative easing is about increasing the quantity of base money. And I believe it might be necessary to purchase longer term and riskier assets than T-bills. And that will directly lower the nominal yields on those assets. Even with an interest rate focus (rather than a quantity of money focus) this is a committment to do what it takes to get nominal expenditure on target.
22. December 2009 at 18:04
Did you see Mark’s latest posting, commenting on a contention by certain economists that monetary policy should pursue level targets instead of rate-of-change targets? And that monetary policy, to work effectively, must be based on setting expectations?
These economists work for… the Federal Reserve. Maybe your ideas are getting through.
Article here
The BBC asked me to come in today to comment on today’s GDP figures, so I took the opportunity to suggest that monetary policy and expectations need to lead us out of the current slow-growth stasis. I wanted to highlight my optimism that this would actually happen, since any policy on which you and Paul Krugman agree must surely represent an emergent consensus of all economists and policymakers in the developed world. But I didn’t have time.
22. December 2009 at 18:48
Doc Merlin, I think you are assuming monetary policy influences AS, when in fact it shifts the AD curve. Labor is by far the most important input, and wages are probably stickier than prices. In any case, they are certainly stickier than NGDP.
Bill, I noticed the same thing, he seemed to equate QE and inflation targeting, but they are two distinct policies (although I suppose they could be combined.)
And of course I agree that it helps to think in terms of nominal expenditure, not inflation. I have a new post on that.
Thanks Leigh. Your comment triggered a new post. BTW, I may visit Oxford in May. Do you live in the UK?
22. December 2009 at 18:49
If you increase the target and expectations are anchored (they equal the target), you’ll get more inflation but you won’t get (much) more output. Perhaps we need more AD, but it doesn’t follow that we need to increase the target.
That said, inflation expectations are about where they were before last Fall. If money policy is too tight, then why aren’t the markets picking up on it?
22. December 2009 at 19:00
pushmdeia1, The SRAS is pretty flat right now. If you get more inflation, output would rise sharply.
Markets expect 1% inflation over the next two years. That’s too little even in normal times. But since mid-2008 inflation has slowed to below trend. We should be doing “level targeting,” which means we need above 2% inflation to “catch up.”
22. December 2009 at 19:02
I’m talking about a shift in the SRAS.
What data are you looking at for expectations?
22. December 2009 at 19:14
Pushmedia1, If you are worried about a shift in SRAS, then target NGDP (the subject of my next post.)
But if you are targeting inflation, and you say perhaps we need more AD, that is equivalent to saying perhaps we need more inflation.
22. December 2009 at 19:41
@ssumner, can you say a few words about why we would need/want to catch up as opposed to just setting forward expectations?
22. December 2009 at 20:11
Professor, you can get more AD without changing the target (e.g. reduce rates or QE)… I think we’re talking about different things. This makes me think that Bernanke was answering a different question than everyone is assuming. You can separately argue policy is too tight but the target is at the right level. He wasn’t talking about the tightness of policy.
If expectations are anchored, you can’t talk about changing the target without talking about moving the SRAS.
22. December 2009 at 21:52
Scott, normally, I agree that monetary policy normally influences AD, however let me lay out my argument for you.
Right now though, going straight to banks what would it do? It wouldn’t push down the rates much, as the 30 year mortgage rates are already about 4.5% and the prime rate is around 3.25% (lowest its been in half a century according to the WSJ). With default rates as high as they are now, lending money at that rate to individuals doesn’t make much sense.
The other option for banks is to buy securities assets. This pushes up AD but weighted heavily towards production inputs other than wages, because the price of consumer goods/finished goods isn’t increasing as fast.
What I described above is very roughly what happened in early 2008. One good way to approximately look at it, is to look at CPIs versus PPIs, but this only works if you have a, more or less, free market. If PPIs are rising faster than CPIs, firms are going to have to cut production. If CPIs are rising faster than PPIs then firms are going to increase production.
This is also a reason that fiscal spending stimulus doesn’t seem to work, while fiscal tax cut stimulus does.
Caveats:
When interest rates are high (or default rates really low), this doesn’t happen, because as the rates drop, home refinance allows for the money to go to consumers very quickly.
22. December 2009 at 22:37
Stockholm Syndrome seems to be an epidemic in Washington.
22. December 2009 at 23:04
“After all, it is spending we are trying to boost, not inflation. Any suggestions?”
Whom do you want to spend?
On what?
When?
If you can answer those questions, you can create incentives and disincentives to try and get the spending you want. I get confused trying to figure out the answers to these questions for some policy proposals.
23. December 2009 at 05:31
“Whom do you want to spend?
On what?
When?”
Anyone who is sitting on cash.
On whatever they want to spend.
Now.
That is why inflation is the right incentive (specially for the now part). Inflation is a tax on cash balanced which induces people to spend today rather than tomorrow.
Alex.
23. December 2009 at 05:42
q, The next post links to an article that discusses this. Here are two advantages to “level targeting”
1. It allows faster recovery from the recession (as wages are very sticky in a downward direction.)
2. Level targeting would make the next recession more mild. This is because if investors expected above normal inflation after an initial period of deflation, they would bid up prices, making the initial deflation less severe.
Pushmedia1, I was discussing a case where the Fed is actually targeting the forecast. A QE or lower interest rate policy that is expected to be successful, will also be expected to cause higher inflation. So if the Fed doesn’t want higher expected inflation, they should not do any kind of expansionary policy.
Doc Merlin. You said;
“Right now though, going straight to banks what would it do?”
What would go straight to banks? I am not talking about helping banking, I am talking about an expansionary monetary policy. The same impact would occur if the Fed bought the securities from the public, and put a penalty rate on ERs to prevent the funds from going straight to banks.
You said;
“It wouldn’t push down the rates much, as the 30 year mortgage rates are already about 4.5% and the prime rate is around 3.25% (lowest its been in half a century according to the WSJ). With default rates as high as they are now, lending money at that rate to individuals doesn’t make much sense.
The other option for banks is to buy securities assets. This pushes up AD but weighted heavily towards production inputs other than wages, because the price of consumer goods/finished goods isn’t increasing as fast.”
The goal is not to get lower interest rates. Low interest rates are an indication of a weak economy, of overly tight monetary policy. In general interest rates are a useless indicator of whether money is tight.
Higher asset prices are a much better way of looking at the picture. If the price of commodities, real estate and stocks rise, then we will get more production of commodities, real estate, and corporate investment. When you say higher commodity prices would reduce final output production, you are confusing a shift in supply with a movement along a supply curve. And expansionary monetary policy will move you up and the the right along a commodity supply curve.
In early 2008 monetary policy was not expansionary, NGDP grew at less that 5%, indeed less than 3%.
Don the libertarian Democrat, I want the American people, and foreigners buying our exports, to spend money on whatever goods they choose. The last thing I want to see is the federal government trying to direct spending. Over the last decade they have been trying to encourage sub-prime mortgages–how did that work out?
23. December 2009 at 06:04
Alex, We think alike, I was posting at the same time.
23. December 2009 at 06:25
“I used to argue that every American would need to have their wage cut 8% relative to a trend line from mid-2008 to restore full employment at current levels of NGDP. I’ve never met anyone who has had that big a wage cut.”
You need to get out more. Bonuses are a part of wages, and most bonuses are a lot smaller than in years past. Had the economy not started a recovery, I would have seen my wage cut by 2/3rd’s relative to 2006. Instead, it’s only down 50%. That is not unusual in my industry, and it is not unusual among business owners whose wages are a function of profits.
btw, I like reading your posts. I agree that a little bit of inflation now would be a good thing.
23. December 2009 at 06:47
Dan Carroll, I’m glad you asked me that. I was talking about hourly wages rates. It is precisely because hourly wage rates have fallen by less that 8% below trend, that other forms of income must fall by more. These include:
1. Profits
2. Bonuses
3. Hours worked
Total national income is now more than 8% below trend. Since hourly wage rates have fallen very little, other forms of income have been hit hard.
And you are right, I do need to get out more.
23. December 2009 at 06:51
“Does anyone know of a good proxy for total spending on all goods and services in the economy?”
It is called Job Guarantee (JG) program.
http://neweconomicperspectives.blogspot.com/2009/11/when-all-else-has-failed-why-not-try.html
http://www.newdeal20.org/?p=6239
23. December 2009 at 06:53
How does #2 work to increase AD in a country with private debt levels at over 200% of GDP? I realize you don’t mean that as a serious proposal, but even on a theoretical level that’s silly.
You’d be on more solid ground just declaring a jubilee on all long term debt.
23. December 2009 at 06:54
Thoma = goldilocks
Monetary policy = oat meal
Papa Bear = Sumner et al
Mama Bear = Inflation hawks
Baby Bear = Bernanke
“This oatmeal is just right”
How does that story end again? 🙂
23. December 2009 at 10:20
‘I used to argue that every American would need to have their wage cut 8% relative to a trend line from mid-2008 to restore full employment at current levels of NGDP.’
– Gee, what a sheltered life….welcome to my world…90 percent..by the way, if everyone takes a wage cut, how does that lead to full employment?
23. December 2009 at 10:33
Regarding Q’s question – the typical arguments (including the ones just made) focus a lot on sticky wages. Not on the stickiness of debt.
Consider the following – with a rate (non-level) target, a one time shock becomes a permanent wealth shock. Consumers react to this rebuilding net wealth and especially liquid wealth (“balance sheet repair”), which means cutting expenditures. Cutting wages fixes the “overemployment” aspect of this, but not necessarily the debt/net wealth aspect (and therefore the aggregate demand aspect) of this. This is where the arguments about wage cuts/increases come into play. While data suggest that arbitrary rage increases are harmful because the harm to unemployment is greater than the benefit from increasing incomes, the evidence is less clear that wage CUTS in an environment of high debt/deleveraging/demand crisis is a GOOD thing. Indeed, one of the primary aspects of the Great Depression was that wages WERE un-stuck… Likewise, with the current/recent recession. Evidence that dropping (nominal) wages even faster would have helped?
When is Macro going to directly engage the net wealth/nominal debt issue? (Also: Why is this perfectly obvious issue so often ignored?)
23. December 2009 at 10:40
Since August 15, 1971, when President Nixon closed the gold window and the world went on a fiat regime, the rules regarding government finance changed completely. Now a sovereign government that is the monopoly provider of a non-convertible floating fx currency of issue is not financially constrained. It does not need to borrow or tax in order to finance its spending, including servicing the national debt. It’s only constraint is real in that if nominal aggregate demand exceeds real output potential, then monetary inflation will rise, but not otherwise. (Milton Friedman: Inflation is always and evertywhere a monetary phenomenon.) When the economy approaches capacity, then the government must either cut spending or raise taxes to withdraw net financial assets from the economy in order to prvent inflation.
Only the federal government as currency issuer can expand and contract the net financial assets of non-government. Commercial banks cannot increase of decrease net financial assets since all their transactions net to zero. This means that when the public desires to save, especially when the country is running a CAD, and business investment is insufficient to fill the gap, then an output gap will result, along with rising unemployment. Unless the government steps in to add net financial assets sufficient to close the gap, there will be an output gap and rising unemployment, and the economy will contract.
What about asset bubbles forming, so that prices in some sectors get out of line with values? Use tageted taxation to take money out of areas that are overheating even in an economy that is otherwise underperforming.
And yes, if a government is profligate, then its currency can decline relative to others, but that gives it an export advantage, too. (BTW, anyone who is afraid that the dollar is going to zero anytime soon can send them to me. I’ll pay the postage.)
Concern about deficits impacting the economy independently of other conditions is misplaced. Deficits are neither good or bad, too big or too small independently of other factors.
The notion that sovereign government that are currency issuers in a non-convertible floating rate regime can go bankrupt, run out of money, become insolvent, or default on their debt is like thinking that scoreboards are limited in the points that they have available. This thinking is a holdover from the previous convertible fixed rate regime, and it is no longer applicable because that’s just not how the monetary system works.
(PELLEY) “Is that tax money that the Fed is spending?”
(BERNANKE) “It’s not tax money. The banks have- accounts with the Fed, much the same way that you have an account in a commercial bank. So, to lend to a bank, we simply use the computer to mark up the size of the account that they have with the Fed.”
When the government wants to spend, the Fed simple debits the Treasury’s reserve (deposit) account. Government borrowing simply clears excess reserves so that the Fed can hit its target (FFR) rate, and there are other ways it could do this, so borrowing is not even necessary at all. Balancing spending with $ for $ borrowing is a voluntary constraint imposed on the system politically that has nothing to do with how the system actually works in practice.
Regarding the US, all this dance about “unsustainable” deficits, “unfunded obligations,” sovereign default, etc, is just kabuki detached from reality. However, it should be noted that the countries of the European Economic Community (EEC) gave up monetary sovereignty, so they can individually default on their debt if not bailed out by the European Central Bank (ECB). US states are in the boat with respect to the federal government as currency issuer, since they are currency users, like households and firms. (Of course, they could use IOU’s as state money if they accepted them in payment of state taxes.)
23. December 2009 at 13:02
If you want people to spend, then why not try:
1) A Dated Coupon ( Incentive to spend now, plus govt borrowing )
2) A Sales Tax Holiday ( Incentive to spend now, plus govt borrowing, plus aid to states )
3) A Tax Break for Investment ( Incentive to spend now, plus govt borrowing )
4) Tax Break for Hiring ( An incentive to spend now, plus govt borrowing )
The govt borrowing leads to expectations of future inflation, which attacks the Flight to Safety, which is the real culprit, including on hiring.
QE, whatever monetary device you use, should:
1) Raise prices over time ( Attacks Deflation which is an incentive for Flight to Safety )
2) Keep Short Term Interest Rates low, as a disincentive to the Flight to Safety.
3) Raise Longer Term Interest Rates, as an incentive for buying corp bonds which fund businesses and attacks the Flight to Safety.
Real World Stories:
http://www.nytimes.com/2009/12/23/business/economy/23views.html?_r=1&src=twt&twt=nytimesbusiness
“Unfortunately for investors, government-related debt isn’t enticing, especially amid growing concerns about national finances. Even the 10-year Treasury note, for example, yields just 3.6 percent annually. Shorter-term paper pays much less.”
http://www.ft.com/cms/s/0/8eeb952a-ee5b-11de-944c-00144feab49a.html?ftcamp=rss
“The Japanese government is easing the tax burden on non-resident investors in Japanese bonds in an effort to boost foreign participation in Japan’s Y933,000bn ($10,300bn) government and corporate bond market. ”
“An official at the Financial Services Agency said that if there were more foreign investors willing to invest in Japanese corporate bonds, this would make it easier for companies with lower ratings to issue bonds.”
http://www.ft.com/cms/s/0/463f62be-edae-11de-ba12-00144feab49a.html
“Total issuance so far this year of high-yield bonds, commonly known as junk, last week topped $144bn (£89bn), passing the previous high of $143bn reached in 2006, according to Dealogic. This marks a big reversal from last year, when investors shunned this corner of the credit markets.”
I could go on.
23. December 2009 at 19:47
Economist, I think you misunderstood the term “proxy” it refers to a variable that captures total spending. Programs aren’t variables.
A payroll tax might help, but expansionary monetary policy would be far more effective. New deal jobs programs didn’t help in the 1930s.
OGT, You are confusing wage rates with income. I am not proposing that we cut the incomes of Americans. So they would still have just as much income to repay their debts after the wage cuts (assuming a given level of NGDP.)
Thruth, I forgot, it’s been 50 years.
Mike, No one in American has had their hourly wage rate cut by 90%, at least for the same job. Most likely you are confusing wage rates with income. Sure, lots of people have seen their incomes fall 90%–it’s called unemployment.
I guess there is a lot of confusion about the concept of wage stickiness. It refers to the wage rate per hour, not profits, bonuses, or changing number of hours worked.
Statsguy, Again, like all the others you seem to confuse wage rates with incomes. Wage cuts do not reduce the income of the average American, and do not increase debt defaults.
What is true is that with nominal debts a policy of NGDP falling is a disaster. And it does (as we’ve seen) greatly increase debt defaults. But for any given fall in NGDP, we are far better of with wage rates falling by the same percentage.
BTW, I am not recommending wage cuts as a policy, it would never work for political reasons. The only reason I included number two was to scare people into favoring an expansionary monetary policy. To embark on a policy of wage cuts to restore full employment is utter madness.
Tom Hickey, Interesting, but how do your ideas relate to my post?
Don, Those second best policies may help a bit, but many will increase the deficit. It would be better to simple expand the money supply enough so the other policies aren’t needed.
You quoted:
“Unfortunately for investors, government-related debt isn’t enticing, especially amid growing concerns about national finances. Even the 10-year Treasury note, for example, yields just 3.6 percent annually. Shorter-term paper pays much less.”
This is an odd quotation, as most economists assume that low yields on a bond are evidence that the bond is very popular, or in high demand. So 10 year bonds must be extremely enticing to yield so little.
In general, monetary policy can only hit one target. So your last three suggestions cannot all be met simultaneously.
23. December 2009 at 20:14
“In general, monetary policy can only hit one target. So your last three suggestions cannot all be met simultaneously.”
Agreed Scott, it also amuses me that when something is targeted it becomes less and less relevant to what the target-er really wants. All in all, I see this as a good thing. It gives them less power/chance to screw up.
23. December 2009 at 20:26
@Scott:
forgot to mention:
“This is an odd quotation, as most economists assume that low yields on a bond are evidence that the bond is very popular, or in high demand. So 10 year bonds must be extremely enticing to yield so little.”
This is partially true, the t-bill being so incredibly low yield is due to four things:
1. The banking regulations making t-bills one of the few things that banks can buy and count same as cash for their reserves.
2. The fed buying up huge numbers of t-bills as part of monetary policy
3. Foreign governments trying to prop up the value of the dollar, for trade balance reasons.
4. Global aging means that people are looking for “safe assets” rather than investments to park their money in (hence the housing fiasco), and the US is still the “tallest pigmy” when it comes to sovereign debt.
24. December 2009 at 06:04
Sumner- Even if one accepts that reducing wages merely redistributes income from workers to employers, you’ve done nothing to address the fundamental problem, which is not enough NGDP for our current nominal debt load. First choice is to increase NGDP, second decrease debt.
I’d argue that there are other reasons adjusting wages down in the current environment wouldn’t be economically wise, such as deflation expectations, debt distribution, and, yes, marginal propensity to spend. But those are all secondary to the fundamental equation, NGDP/Debt.
24. December 2009 at 07:51
Doc Merlin, Your first point is related to “Goodhart’s Law”
I agree with your second comment.
OGT, If wages were cut as much as NGDP has fallen (which I agree is not realistic), then the economy would return to full employment. This is the idea behind the neutrality of money, the idea that if when NGDP fell 10%, all wages and prices fell 10%, then output should not change.
In addition this would not make the debt problem worse.
But I think we both agree that increasing NGDP is far superior, not only does it not hurt the debt situation, it actually helps it.
Whether wage cuts would reduce NGDP is debatable. Krugman says yes, but I doubt it.
24. December 2009 at 09:17
ssumner:
“OGT, If wages were cut as much as NGDP has fallen (which I agree is not realistic), then the economy would return to full employment. This is the idea behind the neutrality of money, the idea that if when NGDP fell 10%, all wages and prices fell 10%, then output should not change.”
I’m going to politely challenge this, but admittedly the wage issue is the one I least well understand of all the macro issues.
You wrote above that wage cuts don’t decrease the income of the _average_ american. I would like to posit the following: the _average_ is not the problem. It’s the distribution that is the problem. In a sense, the same is true with debt – debt is a distributional issue. Bankruptcy/solvency is a distributional issue.
Aggregation up from heterogenous parts yields a well known set of biases in many areas of statistics. So here is the conjecture:
If us assume (briefly) a two part economy – a group of households with little debt and decent income. And a group of households with high debt (and little or no savings), such that debt payments are very close to income.
In such an environment, if you implement a wage cut, the result may very well be that on _average_ there is no decrease, but the distribution of that decrease is such that all Type 2 households go bankrupt (or, more accurately, are set on a highly distruptive and costly bankruptcy trajectory). In a sense, the “stickiness” is because households resist bankruptcy, and there are significant transaction costs associated with liquidation. (Indeed, one can view this in the context of the housing market.)
In such a situation, it is NOT clear that a wage cut is beneficial – there is a strong threshhold effect.
One could argue that if this were true, then we might well expect a wage _increase_ to avoid insolvency, but you’ve argued (and I think convincingly) that this is unlikely based on empirical evidence. Yet I do not think it follows that an arbitrary (govt. mandated – if it were possible) wage cut would necessarily make things better… There are two reasons…
First, the labor response curve could be kinked – in a sense, employers respond to a wage increase by cutting labor, but they do NOT respond to a wage decrease by hiring more labor (certainly, not very rapidly – hiring is itself a sticky process).
Second, the nominal debt threshold effect combined with wealth/income/debt heteroegeneity. (type 1 vs. type 2 households)
What I write is surely blasphemy, so let me end with a simple thought experiment:
What would happen if, in the United States right now, the government were to magically pass a law that forced ALL wage earners to take a uniform 20% price cut?
What does your model say would happen (regarding the unemployment rate and aggregate demand)?
What do you REALLY think would happen (assuming such a law could actually be passed and implemented)?
24. December 2009 at 09:20
Probably not much point in belaboring the point since we are in rough agreement on the practical policy options.
But, even holding wage cuts NGDP neutral, I think the money neutrality theory is the issue where we have different conceptions of the economics. In an economy with a high nominal debt load I am not convinced that it holds for the short or medium term. Least of all in an economy with high nominal debt and a fragile financial system.
It reminds me of AIG, theoretically derivatives should cancel each other out, but if a weak link breaks in the chain…
24. December 2009 at 10:55
I got a 10% pay cut last year.
26. December 2009 at 16:33
Statsguy, I think the distributional issues associated with wages actually favor my argument. Lower wages increase employment, which makes the overall distribution of wage income more equal. Banks are going to see more defaults if 10% of workers lose 100% of pay as compared to a situation where 100% of workers get a 10% pay cut.
If everyone took a 20% pay cut, and assuming the Fed continued to target inflation at around 0% to 2%, then corporate profits would explode upward, more than doubling, as would the stock market. There would be a boom in the economy as investment increased rapidly. This is because the wage cut would overcome the problem of many workers earning more than the opportunity cost of their labor. This may seem odd, but that is only because we can’t imagine this sort of thing because nothing like it has every happened.
BTW, I think you misunderstood my answer to OGT. I assumed both wages and prices fell 10%, in that case it is a tautology that if NGDP falls 10%, output doesn’t change. I was simply describing money neutrality, not trying to prove it.
OGT, Derivitives do balance out. The problem we faced last year was not derivatives, it was the high levels of defaults on mortgages and then later other bank loans. The decision to bail out AIG was a political decision to bail out its creditors.
Again, I do think falling NGDP can cause more defaults, but for any given fall in NGDP I deny that lower wages mean even more defaults that otherwise.
bgp, I assume I don’t know you (although perhaps I do–are you at Bentley?) But very few workers saw their hourly wage rates fall by 10%.